br In this study severe complications on the
In this study, severe complications on the day of treat-ment were rare. Adverse events in the follow-up Bafilomycin-A1 occurred in 36% of patients. Abdominal pain and fatigue were the most common severe (grade 3) adverse events. Clinically important events, such as RILD, were very rare or not reported at all in our cohort. Two patients experienced mild (grade 1) RILD 84 days and 194 days after SIRT. These rates are lower than those in the published literature [17,18].
We recently reported that the combination of SIRT with concomitant oxaliplatinefluorouracil (FOLFOX) chemo-therapy in the first-line treatment of liver metastases from CRC  resulted in neutropenia, febrile neutropenia, thrombocytopenia, fatigue and abdominal pain occurring at a significantly greater frequency in the arm receiving SIRT, albeit at a frequency and severity that was expected and medically manageable. This adverse event profile seems to be related to the combination of SIRT with concomitant chemotherapy, as in the study reported here, in which 65% patients did not receive chemotherapy with SIRT in the salvage setting, severe complications were far less common.
A critical factor in deciding how SIRT should be used in the salvage setting is patient selection. Important sub-groups have been identified in this study that can inform treatment discussions with patients. Patients with no extrahepatic metastases, fewer than six tumours and a tumour-to-liver volume percentage of less than 25% seemed to do better with SIRT, although definite conclusions cannot be drawn without a comparator group. In the recently re-ported first-line studies , an exploratory subgroup analysis showed that patients with liver metastases from right-sided primary CRC may benefit from the combination of SIRT plus concomitant FOLFOX chemotherapy more than patients with liver metastases from left-sided primaries. At the time of designing the registry for the study reported here, this information was not known, so the location of the primary tumour was not one of the fields included in the registry.
Despite the limitations of the registry-based approach, this study shows that this approach to data collection in the health service can accrue rapidly and provide clinically meaningful data. The study has confirmed that SIRT is safe and well tolerated in patients who have previously received multiple lines of chemotherapy and it has shown that SIRT in this population results in overall survival, PFS and LPFS genera are consistent with previously published smaller studies. This study shows the value of a registry-based commissioning model with a systematic research evalua-tion to aid national commissioning decisions for a highly specialist cancer treatment.
Conflicts of Interest
RAS is funded by the NIHR University College London Hospitals Biomedical Research Centre, Cancer Research UK (grant A8971 CRUK/07/030), the CRUK UCL Experimental Medicines Centre and research grants from Sirtex Medical and BTG plc. RAS declares consultancy with Affidea, Astra Zeneca, Boston Scientific, BTG, Cancer Research Technology, DeepMind, Eisai, Sirtex, Terumo and Varian. JKB has received lecturing and consultancy honoraria from BTG and Sirtex Medical. DMM has received honoraria from BTG and Sirtex Medical for sitting on advisory boards.
The authors sincerely thank those patients who took part in the study, the staff at all 10 centres involved and all members of the SIRT CtE Data Working Group for ongoing support and advice in relation to management, design, data collection and analysis. The authors thank Dr Rob Palmer (Cedar, Cardiff & Vale UHB) and Dr Nick Longford (SNTL) for statistical support. We thank the British Society of Inter-ventional Radiology (particularly Dr Fiona Miller, Dr Gra-ham Munneke and Barbara Fletcher) for assistance with the registry and staff at Conexsys Communications Ltd for their contribution to designing and maintaining the SIRT registry. Procedures and data collection were funded by NHS En-gland. NICE was commissioned by NHS England to under-take an independent evaluation. Cedar is funded by NICE as an external assessment centre. Cedar’s work on the SIRT CtE project was funded entirely through a contract with NICE. The SIRT registry was funded by Sirtex Medical.
Appendix A. Supplementary data
 Giammarile F, Bodei L, Chiesa C, Flux G, Forrer F, Kraeber-Bodere F, et al. EANM procedure guideline for the treatment of liver cancer and liver metastases with intra-arterial radioac-tive compounds. Eur J Nucl Med Mol Imaging 2011;38(7): 1393e1406.
 Nicolay NH, Berry DP, Sharma RA. Liver metastases from colorectal cancer: radioembolization with systemic therapy. Nat Rev Clin Oncol 2009;6(12):687e697.